Regularising LSTM classifier by transfer learning for detecting misogynistic tweets with small training set.
Md. Abul BasharRichi NayakNicolas SuzorPublished in: Knowl. Inf. Syst. (2020)
Keyphrases
- transfer learning
- training set
- active learning
- training data
- labeled data
- unlabeled data
- learning tasks
- knowledge transfer
- training samples
- classification algorithm
- training examples
- test set
- class labels
- supervised learning
- learning algorithm
- feature space
- nearest neighbor
- training and test data
- reinforcement learning
- test data
- semi supervised learning
- transfer knowledge
- svm classifier
- cross domain
- data sets
- decision trees
- machine learning
- text classification
- support vector machine
- classification accuracy
- domain adaptation
- target domain
- text categorization
- labeled examples
- collaborative filtering
- cross validation
- decision boundary
- multi task
- machine learning algorithms
- co training
- sample selection bias
- manifold alignment
- text mining
- bayesian networks
- class distribution
- named entities
- feature set
- high dimensional
- data mining